The Study On The Application Of Electronic Nose In The Peanut Ouality Detection

When the quality of the peanut decreased with the increase of storage time, its volatile components will change significantly. The odor fingerprint data of sample can be obtained by detecting with the electronic nose system based on gas sensitive sensors array. Then the quality of the sample can be obtained by the data.The main purpose of this study is to validate the feasibility of non-destructive testing with peanuts by using electronic nose technology. The electronic nose data obtained in this study is used to predict the quality of peanuts. In the research, all samples below were detected by an electronic nose a FOX4000(Alfa MOS, Toulouse, France) with a sensor array of18metal oxide sensors. The samples contained the refreshed peanut and the peanut collected from2010,2011. The physical and chemical quality of each sample was measured with the methods referred by Chinese National Standard.In different conditions (different detection time, space, sample quantity, space generation time and the mode of sample injection), samples in different quality were repeated detected by electronic nose, and the optimal parameters were:180s,200ml,26g,30min, manual injection.The results indicated that there were some similar trends of the sensor data obtained by detecting peanut with and without pods using electronic nose. The peanut of different storage time can be decteted with pods by using electronic nose. Compared with the correlation between physicochemical index and storage time, there also is a good linear relationship between electronic nose data and storage time. The results obtained by analyzing the sensor raw data were better than analyzing the sensor processing data.The PCA and LDA results of detecting the different storage time peanuts with broken pods were better than the CA results. It is feasible that classifying the same storage time peanuts with broken pods and integrity pods by using electronic nose, and it is also feasible to classify the peanut samples with integrity pods and the samples which include the peanuts with broken pods. Besides, the peanuts with different qualitis can be classified by analyzing the sensor data obtained by detecting the peanuts with broken pods or integrith pods.